Features
- Compares various designs of clinical trials
- Uses results from the optimum design of experiments to create powerful and flexible adaptive designs
- Gives examples of clinical trials
- Contains simulations of many procedures for small to moderate-sized samples
- Presents the theory of optimum experimental designs in the appendix
Summary
Randomised Response-Adaptive Designs in Clinical Trials presents methods for the randomised allocation of treatments to patients in sequential clinical trials. Emphasizing the practical application of clinical trial designs, the book is designed for medical and applied statisticians, clinicians, and statisticians in training.
After introducing clinical trials in drug development, the authors assess a simple adaptive design for binary responses without covariates. They discuss randomisation and covariate balance in normally distributed responses and cover many important response-adaptive designs for binary responses. The book then develops response-adaptive designs for continuous and longitudinal responses, optimum designs with covariates, and response-adaptive designs with covariates. It also covers response-adaptive designs that are derived by optimising an objective function subject to constraints on the variance of estimated parametric functions. The concluding chapter explores future directions in the development of adaptive designs.
Table of Contents
Introduction: Stories and Data
Scope and Limits
Two-Treatment Trials with a Binary Response
Equal Randomisation
Adaptive Allocation
Urn Model
Some Motivating Clinical Trials
Adaptive Design: Controversies and Progress
Why Adaptive?
How Adaptive?
Criticism
What Next?
Randomised Balanced Sequential Treatment Allocation
Introduction
Balance with Two Treatments
Designs with Three or More Treatments
Designs with Covariates
The Distribution of Loss and of Bias
Heteroscedastic Models
More about Biased-Coin Designs
Further Reading
Response-Adaptive Designs for Binary Responses
Introduction
Urn Designs
Play-the-Winner Rule
Randomised Play-the-Winner Rule
Generalised Pólya Urn
Success Driven Design (SDD)
Failure-Driven Design (FDD)
Birth and Death Urn (BDU)
Birth and Death Urn with Immigration
Drop-the-Loser Rule
Odds Ratio-Based Adaptive Designs
Delayed Response in the RPW Rule
Prognostic Factors in Urn Designs
Targeting an Allocation Proportion
Adaptive Designs for Categorical Responses
Comparisons and Recommendations
Response-Adaptive Designs for Continuous Responses
Motivation
Some Trials with Continuous Responses
Doubly Adaptive Biased-Coin Designs (DBCD)
Nonparametric Designs
Adaptive Designs for Survival Data
Link Function-Based Adaptive Design (BB)
Multi-Treatment Multivariate Design
DL Rule for Continuous Responses (CDL)
Response-Adaptive Designs for Longitudinal Responses
Repeated Responses
Binary Longitudinal Responses (SLPW)
Design and Analysis for the PEMF Data
Longitudinal Categorical Responses
Longitudinal Multivariate Ordinal Responses
Models with Covariates
Continuous Longitudinal Responses
Random Number of Responses
Numerical Illustrations
Optimum Biased-Coin Designs with Covariates
Modelling and Design
Biased-Coin DA-Optimum Designs
Numerical Comparisons for Two Treatments
Designs for Three Treatments
Distribution of Loss
Skewed Allocations
Skewed Allocation – Numerical
Heteroscedastic Normal Models
Allocation Rules for Heteroscedastic Models
Generalized Linear Models
Binary Data
Allocation Rules for Binomial Models
Gamma Data
Loss, Power, Variability
Further Reading: Skewed Designs
Optimum Response-Adaptive Designs with Covariates
Introduction
Link-Function-Based Adaptive Design
Adaptive Designs Maximising Utility
Power Comparisons for Four Rules
Redesigning a Trial: Fluoxetine Hydrochloride
Extensions
Further Reading
Optimal Response-Adaptive Designs with Constraints
Optimal Designs Subject to Constraints
Design of Jennison and Turnbull
RSIHR Design
Maximising Power: Neyman Allocation
Other Designs
BM Design
ZR Design
A General Framework: BBZ Design
Two Normal Populations with Unknown Variances
Two-Sample Nonparametric Design
BM Design for More Than Two Treatments
Optimal Designs with More than One Constraint
Designs for Survival Times
Covariates
Implementation
Adaptive Constraints
Back to Chapter 7
Adaptive Design: Further Important Issues
Bayesian Adaptive Designs
Two-Stage Adaptive Design
Group Sequential Adaptive Design
Optimal Design for Binary Longitudinal Responses
Inverse Sampling
Robustness in Adaptive Designs
Missing Data in Response-Adaptive Designs
Asymptotic Results for CARA Designs
How to Bridge Theory and Practice
Appendix: Optimum Design
Bibliography
Index
Reviews
"… an excellent textbook on this important topic. In general, the book is well written, easy to navigate, and definitely consistent with the high quality of other books in Monographs on Statistics and Applied Probability series by CRC Press. … a well-structured and clearly presented textbook on randomized response-adaptive designs. … Because the book offers a well-balanced mix of practical applications and theoretical results, a wide range of readers, from graduate students to applied statisticians with solid mathematical background, will find the book useful. Readers may find the authors’ emphasis on the use of simulation methods to compare methods particularly useful."
—The American Statistician, August 2015
"Atkinson and Biswas address [the] questions throughout in a thorough and interesting manner via a logical structure that makes navigating the text a simple and intuitive exercise. … ideal for anyone looking to learn more about the growing field of response adaptive designs. An easy read, it is well written from the off set, logically advancing the mathematical complexity as the chapters proceed, making it useful for those new to the field of clinical trial design and seasoned trial statisticians alike. Through its use of numerous examples, it clearly achieves its stated aim; to elucidate the practical usefulness of this class of designs. Given their increasing popularity, it may well be one book you should consider adding to your collection sooner rather than later."
—ISCB News, 59, June 2015
"This book is clearly written and well-structured for a graduate course as well as consulting statisticians. … this book is particularly useful for the development of orphan drugs."
—Biometrics, March 2015
"… the book covers a very broad range of response-adaptive designs and related issues. … the book is comprehensively written and gives many exemplary applications from clinical practice. At the same time, the book is mathematically sound and also provides the underlying mathematical formulas and derivations. For these reasons, the book offers important content for applied statisticians but also for more theoretically interested mathematicians."
—Biometrical Journal, 2014
